Gokulakrishnan
Full Stack Engineer Intern @Thebinaryholdings
Overview
Full Stack Engineer Intern @Thebinaryholdings
Founder @Quarix
Social Links
About
I'm a Full Stack Engineer and AI Engineer with 2+ years of freelance experience, and the Founder of Quarix.
I build things at the intersection of intelligent systems and great user experiences — from production-grade web and mobile applications to custom AI agents, LLM integrations, and RAG pipelines.
My core stack spans Python, Next.js, React, and TypeScript, with deep expertise in Machine Learning, LLMs, and modern AI tooling. I don't just integrate AI — I architect it, from fine-tuning models to deploying them at scale.
Before going independent, I sharpened my engineering foundation as a Full Stack Engineer Intern at The Binary Holdings, working on real-world products in a fast-paced environment.
Through Quarix, I partner with startups and businesses to deliver:
- Custom AI Agents & Chatbots
- MCP server development
- LLM integration & fine-tuning
- RAG pipelines & model deployment
I'm driven by one thing: turning ambitious ideas into polished, working products. Whether it's a sleek web app or a complex AI system, I bring the same obsession with craft to everything I build.
GitHub Contributions
Stack
Experience
- Building the largest Web3 distribution infrastructure for telcos.
- Designing and building applications for TBH to improve the services.
- TypeScript
- Next.js
- React
- Tailwind CSS
- Figma
- Business Ownership
- Business Law
- Business Tax
Freelance
- Laravel
- React
- Express.js
- Socket.IO
- MongoDB
- Firebase
- WordPress
- Docker
- NGINX
- Creativity
- UI/UX Design
- Graphic Design
- Sketch
Education
- B.Tech. in Computer Science (Specialization: AI & Data Science)
- CGPA: 8.26/10
- Language Proficiency: English
- Technical Lead of Blue Screen Programming Club (2024-2025)
- Hosted Innothon 2025 CodeArena with 1,000+ participants
- Organized multiple technical events and hands-on workshops for the club
- Drove student engagement and programming skill development
- C++
- Java
- Python
- PHP
- DSA
- Advanced Databases
- Systems Design
- Distributed Systems
- Software Engineering
- Artificial Intelligence
- Data Science
- Machine Learning
- Event Management
- Workshop Organization
- Leadership
- Team Coordination
- Self-learning
- Teamwork
- Leadership
- Sports
- Academics
- Teamwork
Projects(4)
4th Place Winner – OpenAI Academy × NxtWave Buildathon (competed against 1,000+ teams).
- AI-powered health assistant that provides personalized medical guidance using Google Gemini AI.
- Medical records management to organize and analyze all health documents in a single, secure app.
- Emergency doctor discovery with interactive maps for locating nearby doctors and hospitals.
- Medication reminders and adherence monitoring so users never miss a dose.
- Voice interaction via speech recognition and text-to-speech for hands-free health conversations.
- NFC emergency access to critical health data for rapid response in emergency scenarios.
- Flutter
- Dart
- Supabase
- Google Gemini AI
- Google Maps
- NFC
- Mobile
Organization: Central Vision Eye, Chennai, India
- AI-powered predictions using machine learning models trained on 250+ anonymized patient records.
- Visual analytics with interactive charts and graphs for myopia progression and risk assessment.
- PDF clinical reports that summarize risk, treatment recommendations, and patient-friendly explanations.
- Treatment efficacy estimation to quantify projected Stellest lens effectiveness over time.
- Real-time analysis that produces instant predictions from clinician-entered clinical data.
- Next.js
- TypeScript
- Python
- FastAPI
- Machine Learning
- scikit-learn
- Data Visualization
- Tailwind CSS
- shadcn/ui
- Recharts
Recognized with Best Paper Award at IEEE International Conference, Singapore (2025).
3D volumetric mapping
- Utilized a voxel-based occupancy grid (OctoMap) to represent the 3D environment.
- Each drone processes raw LiDAR point clouds in real time to classify space as free, occupied, or unexplored.
Frontier-based navigation
- Implemented a frontier detection algorithm to find boundaries between known and unknown regions.
- Autonomous agents compute optimal trajectories toward frontiers to maximize the rate of map expansion.
Federated learning framework
- Local policy training: each drone trains its own RL policy based on local flight paths and obstacle-avoidance experiences.
- On-device processing keeps training at the edge, avoiding transmission of large raw sensor datasets.
Model aggregation (FedAvg)
- A central federated server periodically collects local model weights (not raw data).
- Implemented FedAvg (Federated Averaging) to fuse individual drone experiences into a shared global policy.
- The updated global model is broadcast back to the swarm, allowing drones to benefit from regions they have not personally explored.
- Reinforcement Learning
- Federated Learning
- FedAvg
- LiDAR
- 3D Mapping
- OctoMap
- Voxel Grid
- Robotics
Freelance project for building a stock portfolio tracking platform.
- Tracked a sample equity portfolio starting with an initial capital of ₹10 Lakhs.
- Collected, cleaned, and structured quarterly stock performance and allocation data.
- Built a web interface to present quarterly picks, profit/loss breakdowns, and cumulative portfolio performance.
- Focused on clear data visualization so investors can quickly understand trends and returns.
- Data Visualization
- R
- Python
- React
